Glomulina Oculus, New Calcareous Foraminiferal Species from the High Arctic: a Potential Indicator of a Nearby Marine-Terminating Glacier

Journal of Foraminiferal Research, v. 50, no. 2, p. 219–234, April 2020

GLOMULINA OCULUS, NEW CALCAREOUS FORAMINIFERAL SPECIES FROM THE HIGH ARCTIC: A POTENTIAL INDICATOR OF A NEARBY MARINE-TERMINATING GLACIER

Anne E. Jennings1,*, Marit-Solveig Seidenkrantz2 and Karen Luise Knudsen2

ABSTRACT Sound. In addition, it has been found in both surface sediments (one Rumohr core top sample) and early Holocene A new calcareous Arctic foraminiferal species, Glomulina glaciomarine sediments from the NE Greenland shelf. oculus n. sp., belonging to the suborder Miliolina has been ob- Herein we describe the new species and its modern envi- served in surface samples from northern Nares Strait and Pe- ronmental preferences as well as the implications of its oc- termann Fjord, NW Greenland, and off Zachariae Isbrae, NE currences in paleoenvironmental reconstructions. We only Greenland, as well as in early Holocene sediments from the include the data relevant for the presentation of this new northern Baffin Bay region and on the NE Greenland shelf. In species, while further investigations of the foraminiferal as- some samples, this new porcelaneous species makes up a sig- semblages in the Lancaster Sound and NE Greenland shelf nificant fraction of the foraminiferal assemblage, particularly study sites will be presented elsewhere. in samples with a relatively large sand content, and we suggest that this species is indicative of an Arctic environment with marine-terminating glaciers. Yet, further studies are needed STUDY REGIONS to ascertain its full habitat range. Northern Nares Strait and Petermann Fjord INTRODUCTION The modern samples for this study were primarily ac- quired from the northern Nares Strait (northern Kennedy A new species of calcareous benthic foraminifera belong- Channel, Hall Basin, southern Robeson Channel and the ing to the suborder Miliolina was found in modern (includ- Petermann Fjord, including beneath the floating ice tongue; ing Rose Bengal stained specimens) and Holocene seafloor Fig. 1A, B). These high Arctic sites are situated above 80°N sediments from Petermann Fjord and northern Nares Strait and are strongly influenced by both sea ice and terrestrial in NW Greenland, in the northernmost Baffin Bay, and on glacier ice, including plateau ice caps and the Petermann the NE Greenland shelf. Despite it being a distinct and com- Glacier, a main outlet glacier of the Greenland ice sheet mon species in these areas, careful searches of the litera- that terminates in a 47-km-long floating ice tongue (Mün- ture and queries to other foraminiferal experts with expe- chow et al., 2016). The area is bathymetrically complex with rience in Arctic faunas did not yield any previous identi- many basins formed by underlying bedrock topography and fication of this species. Therefore, we have concluded that molded by confluent ice streams during the last glaciation the species, though common in certain Arctic environments, (Jakobsson et al., 2018). has not been described previously. The species is referred to The water column in the region is stratified with very cold, the genus Glomulina, bringing the number of species in this low salinity Polar Water from the Arctic Ocean forming the genus to three (cf., Ellis & Messina, 2019) and we here name upper ∼50 to 100 m, overlying relatively warm and saline the species Glomulina oculus. (∼0.3°C, 34.8 psu) Atlantic Water (Fig. 2). The Atlantic Wa- Glomulina oculus n. sp. was first noted in modern and ter is derived from the Atlantic Layer in the Arctic Ocean, Holocene seafloor sediments sampled during the 2015 Peter- and it enters Nares Strait over a 290-m-deep sill in the Lin- mann Glacier Expedition to northern Nares Strait and the coln Sea (Münchow et al., 2011, 2016) before it passes over Petermann Fjord, NW Greenland. The species was noted a 440-m-deep sill at the mouth of the Petermann Fjord to routinely during shipboard scanning for foraminiferal con- form the bottom water in the fjord (Münchow et al., 2016; tent in sediment cores during visual core descriptions. Both Jakobsson et al., 2018; Fig. 1B). Rising glacier meltwater living (Rose Bengal stained) and dead (unstained) specimens and Atlantic Water mix near the Petermann Glacier ground- were observed in many of the multicore tops. Rare unstained ing line to form a transitional layer above the Atlantic Wa- Glomulina oculus specimens were furthermore observed in ter (Münchow et al., 2016; Washam et al., 2019). Slight but the Uwitec core tops from below the floating ice tongue. significant warming of Atlantic Water has been observed in Subsequently, Glomulina oculus was also identified in early Nares Strait and the Petermann Fjord over the last decade Holocene records from marine sediment cores in the north- (Münchow et al., 2011; Washam et al., 2018). The Peter- ernmost Baffin Bay, near Smith Sound and in Lancaster mann Glacier meltwater influx can be traced throughout Hall Basin but is relatively more concentrated along the Greenland side as it exits the fjord (Heuzé et al., 2017). 1 INSTAAR, University of Colorado, Boulder, Colorado 80309-0450 Land-fast sea ice covers the northern Nares Strait for USA mostoftheyear,followedbya3to4-monthperiodof 2 Paleoceanography and Paleoclimate Group, Department of Geo- mobile pack ice. The sea-ice regime is governed by the for- science, Arctic Research Centre, and Climate Interdisciplinary Cen- tre for Climate Change, Aarhus University, Høegh-Guldbergs Gade 2, mation of ice arches that form by consolidation of thick 8000 Aarhus C, Denmark multi-year Arctic Ocean sea-ice floes that converge in the * Correspondence author. E-mail: [email protected] Lincoln Sea at the northern end of the strait (Kozo, 1991),

219

Downloaded from https://pubs.geoscienceworld.org/cushmanfoundation/jfr/article-pdf/50/2/219/5092044/i0096-1191-50-2-219.pdf by University of Colorado Boulder user on 21 July 2020 220 JENNINGS AND OTHERS

Figure 1. A Map depicting the cores (stars), where Glomulina oculus n. sp. has been found in early Holocene glaciomarine sediments in Hall Basin, northernmost Baﬃn Bay, Lancaster Sound, and on the NE Greenland shelf. SS = Smith Sound, LS = Lancaster Sound, PF = Petermann Fjord, HB = Hall Basin, KC = Kennedy Channel, KB = Kane Basin, 05 = core Hly03O5GC, 14 = piston core 2001LSSL-014PC, 49 = piston core 2008029-49PC, 59 = piston core 2008029-59PC, 73G = gravity core DA17-NG-ST07-73G, 90R = Rumohr core DA17-NG-ST08-90R. The box indicates the area shownindetailinFig.1B.B Detailed map of northern Nares Strait and Petermann Fjord showing locations of multicores (white circles) and Uwitec cores (black circles), where modern samples were taken. Specimens of G. oculus were found in all these surface sediment samples.

formation of the North Water Polynya, a large area of thin to absent sea ice that encompasses the core sites in Smith and Lancaster sounds (Dunbar, 1969; Melling et al., 2001; Tang et al., 2004).

NE Greenland Shelf The investigated sites off NE Greenland are today characterized by chilled Atlantic Water returning southward along Greenland as a subsurface water mass that originated in the North Atlantic Drift and either returns southward off Svalbard or as a subsurface flow of Atlantic Water that has circuited the Arctic Ocean (Aagaard & Coachman, 1968a, b). The relatively warmer Atlantic-source waters reached to the glacier tongues of NE Greenland through deep channels (Schaffer et al., 2019). Atlantic Water is overlain by Polar surface water from the Arctic Ocean (East Greenland Cur- rent; Aagaard & Coachman, 1968a, b) combined with lo- cally formed surface water from melting of the Greenland ice sheet. One of our sites is located close to Zachariae Isbrae, while the second site is located farther offshore. Zachariae Isbrae likely extended much further onto the shelf and closer to the study sites during the deglaciation and the early Holocene (Larsen et al., 2018). Figure 2. Conductivity-Temperature-Depth (CTD) profiles from OD1507-36CTD and -15CTD (Fig. 1B) the CTD casts nearest to sites OD1507-23MC and -47MC with the largest numbers of Glomulina ocu- MATERIALS AND METHODS lus. Data from Muenchow (2019). PW = Polar Water, TL = Transi- tional Layer, AW = Atlantic Water. During the Petermann Glacier Expedition of Swedish Icebreaker Oden in 2015, the uppermost 1 to 2 cm of seafloor sediments were collected for foraminiferal analysis or more commonly in the Kane Basin at the southern end. from12multicoresat11sitesaswellasfromthetop2 Ice arches form in late fall and early winter and effectively cm of two modified Uwitec cores raised from beneath the block Arctic sea-ice flow through the Nares Strait (Kwok, floating ice tongue (cf., Makinson & Anker, 2014). The 2005), while allowing Polar Water to flow beneath the sea present water depths ranged from 517 to 1041 m (Fig. 1B), ice. Break-up of the ice arches occurs sometime between the bottom water temperatures between 0.27 to 0.3°C, and mid-July to mid-August, resulting in mobilization of sea ice bottom salinities from 34.77 to 34.82 psu. Sample volumes through Nares Strait toward Baffin Bay driven by strong ranged between 6 and 35 ml. Immediately after sampling northerly winds (Samelson et al., 2006). Sea-ice conditions the multicore tops on deck, a buffered solution comprising in recent years are more variable as the formation of north- 70% distilled water and 30% alcohol, by weight, 1 g baking ern and southern ice arches become less reliable, resulting in soda and 1 g of Rose Bengal (dichlorotetraiodofluorescein) some years with a longer or even continuous mobile sea-ice stain was added to the vials to preserve the foraminifera and season (Kwok et al., 2010; Ryan & Münchow, 2017). to stain living specimens. The samples were immediately refrigerated. In addition to the surface samples, numerous Northern Baffin Bay sediment cores, representing the upper 10 m of the sediment section, were split and described during the cruise. As part The northern Baffin Bay, including Lancaster Sound, sed- of the core-description process, very small samples were iment core sites (Fig. 1A), are influenced by cold, low salin- routinely sieved at 0.063 mm to observe the faunal content ity Polar Water from the surface of the Arctic Ocean that with depth in the cores. The new species, Glomulina oculus, flows through both Lancaster Sound and Nares Strait form- was commonly observed during this process. ing the surface water in Baffin Bay (Tang et al., 2004). At- Foraminiferal analysis of the surface samples was com- lantic Water from the Arctic Ocean is largely excluded by pleted in the Micropaleontology Laboratory at INSTAAR, the shallow sills in these channels. The Atlantic Water that University of Colorado. The stained samples were washed reaches northern Baffin Bay is carried in the West Green- through a 0.063-mm mesh sieve. The >0.063 mm fraction land Current; it submerges beneath the Polar Water when was stored in a buffered solution (70% distilled water and reaching the northern Baffin Bay (Tang et al., 2004). Sea ice 30% alcohol, by weight, and 1 g baking soda) to keep the covers nearly all of Baffin Bay in winter; it begins to form pH >8and<<9. Analysis of these foraminiferal samples was in September and reaches maximum coverage in March. It carried out in the buffered solution in a foraminiferal picking is thickest along Baffin Island, where the flow of the low tray under a binocular microscope. Both living (Rose Bengal salinity Arctic surface water is concentrated. However, the stained) and dead (unstained) specimens were counted and ice arches that form in Nares Strait in most years block pas- are reported as total assemblages herein. In order to de- sage of sea-ice floes from the Arctic Ocean and facilitate the termine whether G. oculus would be preserved using a dry

sample analysis, we also wet-sieved sediment from equiva- lent samples that had been freeze-dried, and subsequently we oven-dried the >0.063-mm fraction at 25°C. Glomulina oculus was found both in wet and freeze-dried samples. Photographs of specimens in the storage solution were taken, and specimens were picked from wet and dry samples using a brush, allowed to dry, stored on cardboard slides and photographed. Specimens picked from the buffered solution were rinsed in distilled water before placing them on slides. The specimens were sent to Aarhus University for scanning electron microscope (SEM) photomicrographs. Here the gold-coated specimens were photographed using a Versa 3D FIB-SEM at the iNANO Cryo-Electron Microscopy Facility. Fossil samples from sediment gravity core Hly03-05GC Figure 3. Glomulina oculus specimens per ml of sediment sam- from Hall Basin (81.621°N, 63.26°W; 797 m water depth) ple compared to the weight percentage of sand at each multicore site. Numbers above the G. oculus/ml bars represent the ratio of living (Rose were prepared by freeze-drying, followed by sieving and then Bengal stained) counts to total counts of G. oculus (note that a value air drying of the 0.063 mm sediment (Jennings et al., 2011). of 1 indicates that all of the specimens were living, while a value of 0 For piston core 2001LSSL-014PC in northern Baffin Bay indicates that all specimens were dead at the time of collection. Sand (77.702°N, 75.07°W; 658 m water depth), the wet procedure percentage data are not available for the Uwitec core tops so the two used for the modern OD1507 samples was used to study sub-ice tongue sites are not included in the figure. the foraminifera (Jennings et al., 2019). The age control for these two cores was based on radiocarbon dating of benthic and planktonic foraminifera, and the resulting age con- analysis with an overall Holocene rate from radiocarbon- trol and the procedures used to generate age models and based analyses of 0.06 cm a−1 (Reilly et al., 2019). There calibrate the ages all are presented in respective publica- is a tendency for the samples with higher percentages of tions. The sand percentages from surface samples in north- sand (>0.063 to <1000 mm) to have higher numbers of spec- ern Nares Strait and northern Baffin Bay were obtained us- imens of G. oculus (Fig. 3). Samples outside of the Peter- ing the Malvern Mastersizer 3000 laser diffraction particle mann Fjord, except for 47MC, tended to have greater sand size analyzer (cf., Jennings & Andrews, 2019; Jennings et al., percentages and greater abundances of G. oculus (Fig. 3). 2019) whereas weight% sand data from the Hly03-05GC Glomulina oculus was present at under 2% of the total (Hall Basin) was obtained by weighing the sieve fractions (calcareous and agglutinated) benthic foraminiferal assem- used for foraminiferal analysis (Jennings et al., 2011). blages, except at two sites in Hall Basin near the mouth Foraminiferal and grain-size samples from the cores in of the Petermann Fjord. At these two stations (multicores northern Baffin Bay (2008029-49PC: 74.026°N, -77.13° W; 47MC and 23MC), G. oculus occurred at 6.4 and 15.4% 868 m water depth; 2008029-59PC: 74.258°N, 82.23°W, 791 of the total benthic foraminiferal assemblage, respectively m water depth; 2001LSSL-14PC were treated with the same (Fig. 4). Figure 4 shows the additional species that occur at methods as those for the OD1507 samples. >5% of the assemblage at these two sites. Elphidium clavatum Samples from gravity core DA17-NG-ST07-73G (Cushman) (cf., Darling et al., 2016) is the most abundant (79°04.100N, 11°54.191W; 385 m water depth) and species, making up 37.7% and 26.5%. Cassidulina neoteretis Rumohr core DA17-NG-ST08-90R (78°30.001N, Seidenkrantz, Nonionella iridea Heron-Allen and Earland, 17°18.431W; 594.9 m water depth) from the Northeast Stetsonia horvathi Green and the agglutinated species Tex- Greenland shelf were prepared at the Department of Geo- tularia earlandi Parker are also common at both sites. Epis- science, Aarhus University, through wet-sieving of the tominella arctica Green makes up 12.3% of the assemblage sample on 0.063-mm and 0.1-mm sieves. After sieving, the in 23MC but is absent in 47MC. The high percentages of E. samples were dried prior to analyses. clavatum are consistent with the outflow of glacial meltwater (Hald & Korsun, 1997, as E. excavatum)fromthePetermann Glacier that is documented to be most prominent along the ECOLOGY AND PALEOENVIRONMENTS Greenland side of Hall Basin (Heuzé et al., 2017; Washam Modern Samples in Northern Nares Strait and off NE et al., 2019). Taken together, the other calcareous species are Greenland consistent with the modern environment of a stratified water column with chilled and modified Atlantic Water underly- Glomulina oculus n. sp. was found in every surface sample ing Polar Water in an environment with seasonally mobile collected in northern Nares Strait and the Petermann Fjord, sea-ice cover that allows episodic marine productivity (e.g., including in surface sediment samples from beneath the ice Wollenburg & Mackensen, 1998; Jennings et al., 2004). tongue, where 1 and 2 dead specimens were found in 006 Glomulina oculus is also present, albeit not abundant, in and 001 UW, respectively (Fig. 1B). It was found at abun- muddy surface sediments from a Rumohr core DA17-NG- dances ranging from close to zero to nearly 50 specimens ST08-90R (594.9 m water depth; Fig. 1A; data not shown) per ml of sediment with ratios of living to total counts rang- taken from the deeper, inner part of a trench extending from ing between 0 and 0.9 (Fig. 3). Sediment accumulation rates the glacier tongue of Zachariae Isbrae to the shelf edge of in Petermann Fjord were estimated at 0.3–1 cm from 210Pb NE Greenland. The dominant species of the assemblage is

Figure 4. Glomulina oculus and other benthic foraminiferal species that occur at >5% in multicore surface samples at 47MC and 23MC outside of the Petermann Fjord (see Fig. 1B for locations).

Cassidulina neoteretis (data not shown). Chilled return At- Arctic Ocean due to isostatic depression immediately after lantic Water is flowing along the sea floor through this deep glacier recession (Jennings et al., 2011), and it also likely re- trench, while surface waters are characterized by Polar Wa- flects low food supply during initial deglaciation (Osterman ter from the East Greenland Current (Aagaard & Coach- et al., 1999). Increasing percentages of E. clavatum and N. iri- man, 1968a, b; Schaffer et al., 2019). The site is normally dea suggest more open water that promoted greater marine covered by perennial sea ice, but unusually low sea-ice cover productivity during deposition of the bioturbated mud that in late September 2017 allowed us to reach this site; the core closely followed opening of the strait (Gooday & Hughes, was taken from right at the sea ice edge off the glacier. 2002; Fig. 6A).

Occurrences of GLOMULINA OCULUS in Holocene Sediment Cores Glomulina oculus n. sp. has been found in several sediment cores, among those are two cores with published foraminiferal faunal data (Jennings et al., 2011, 2019). How- ever, specimens of G. oculus have hitherto been assigned to other taxa. In both previously published cores, the absolute abundances, as well as relative frequency of G. oculus,areat a maximum in or close to intervals with relatively high sand contents that are associated with transitional environments related to deglaciation (Figs. 5, 6). In core Hly03-05GC from Hall Basin (Fig. 1A, B), the region that supports the occurrence of G. oculus today, it was misidentified as Miliolinella subrotunda (Montagu) by Jen- nings et al. (2011) because of its similar appearance to juveniles of this species. In Hly03-05GC, G. oculus was present in very low percentages in several samples of the distal glaciomarine laminated clay that was deposited in front of the confluent Greenland and Innuitian ice sheets (Jennings et al., 2011). However, its percentages increased greatly in the sediments encompassing the transition from a glacial embayment to the open strait that followed ice retreat, but the species did not occur in sediments above this interval (Fig. 6A). In the early Holocene interval between ∼7and 10 cal ka BP, where G. oculus is found, it is associated with C. neoteretis as the most dominant species, suggesting presence of chilled Atlantic Water (e.g., Seidenkrantz, 1995; Jen- nings & Helgadottir, 1994). Oridorsalis umbonatus (Reuss) Figure 5. Glomulina oculus specimens per g of sediment sample increases near the top of the laminated mud unit and con- compared to the weight percentage of sand downcore in A Hly03O5GC tinues into the base of the bioturbated mud unit. Its oc- from Hall Basin, and B in 2001LSSL-14PC from northernmost Baffin currence was interpreted to reflect deeper connection to the Bay. Data from Jennings et al. (2011, 2019).

Figure 6. Downcore percentages of Glomulina oculus and other important species in key intervals of A Hly03O5GC from Hall Basin, and B in 2001LSSL-14PC from northernmost Baﬃn Bay. Data from Jennings et al. (2011, 2019).

Glomulina oculus also was found in sediment core imens have also been identified in early Holocene (ca 8–9 2001LSSL-014PC in northern Baffin Bay, just south of cal ka BP) sediments of core DA17-NG-ST07-073G on the Smith Sound (Fig. 1A; incorrectly named as Cyclogyra sp.), NE Greenland shelf (385 m water depth; data not shown). in the very first fossiliferous samples of glaciomarine sed- These sediments were deposited just above sediments rich iments above a till. Its occurrence was limited to the top in reworked foraminifera from the Pliocene or early Pleis- of a sediment facies of laminated ice proximal glacioma- tocene, indicating a period of sediments being flushed out rine sediments (L2, Jennings et al., 2019) and the base of during the deglacial retreat of the Greenland ice sheet. overlying pebbly mud (L3, Jennings et al., 2019) deposited Given the fact that we have now identified the species in at ∼11.2 cal ka BP during calving margin ice retreat, prior several surface sediment samples and cores from two ge- to the opening of Nares Strait (Fig. 6B). The sedimenta- ographically separate regions, we suspected that G. oculus tion rate in this interval is estimated at 0.13 cm a−1.Across has not been previously reported because it might have low this interval in the core, G. oculus is associated with high preservation potential or be too fragile to survive the com- percentages of S. horvathi, a species common in the Arctic mon practice of drying foraminiferal samples prior to siev- Ocean that is indicative of low food supply and heavy sea- ing. Consequently, we tested its preservation potential in ice cover. As in Hly03-05GC, it is associated with the chilled two ways. First, we sieved previously freeze-dried samples Atlantic Water species C. neoteretis, but it also co-occurs from multicore tops 23MC and 45MC that were known to with the (sub)Arctic species Islandiella norcrossi (Cushman), have relatively high numbers of the species when prepared a species that is indicative of stable bottom salinities and with the wet procedures. We found that G. oculus in these highly chilled Atlantic Water (Lloyd, 2006; Fig. 6B). paired samples not only survived freeze-drying, they also True to its association with transitional glacial marine were present after oven drying at 25°C. In fact, several ex- environments, abundant G. oculus has also been found amples of these dried specimens were picked for SEM imag- in glaciomarine sediments closely overlying a till in two ing. In core Hly03-05GC, the foraminiferal samples were cores from Lancaster Sound (data not shown): 2008029- prepared by freeze-drying followed by sieving at 0.063 mm 49PC (868 m water depth) and -59PC (791 m water depth) and then air-drying of the 0.063 mm sediment fraction. The (Paratype from 59PC illustrated in Figure 7.7). A few spec- presence of G. oculus in these samples shows that the species

is preserved after drying, and its presence/absence does not and 726888) from 2008029-59PC, 549-551 cm (Fig. 7.7). rely on the wet technique. It seems likely that G. oculus has Paratypes 12 and 13 (5 chambers; USNM PAL 726889 and been observed at more sites than we are currently aware of, 726890) from OD1507-34MC-1, 0–2 cm (Fig. 7.6). but that it has been left unidentified or has been assigned Material. Early Holocene to Recent; northern Nares to other genera and species as was the case for the studies Strait, northern Baffin Bay, NE Greenland shelf. of cores Hly03-05GC (Jennings et al., 2011) and 2001LSSL- Type locality. Hall Basin, northern Nares Strait. 014PC (Jennings et al., 2019). Glomulina oculus has not to Type level. Recent; early Holocene. our knowledge been described or illustrated previously. Description. Test discoidal to globular, round to slightly In summary, based on its modern occurrences in north- elongate with 3–4 chambers primarily organized in one ern Nares Strait, Glomulina oculus is a (high) Arctic ben- plane, but with tendency for the third and fourth cham- thic foraminiferal species that lives in areas of perennial bers to be formed slightly off-center from the periphery (i.e., to mobile sea ice beneath a stratified water column with weakly streptospiral). The species has a large, spherical pro- chilled Atlantic Water, overlain by Atlantic Water mixed loculus followed by normally two or three tubular cham- with glacier meltwater, all overlain by Polar Water from the bers. The proloculus is wrapped by the first tubular cham- Arctic Ocean. Elevated abundances of G. oculus are associ- ber of ∼¼ of a whorl in length. The third chamber makes up ated with increased sand content, probably originating from ∼¾ of a whorl with a slight change in the coiling plane (i.e., sea ice and iceberg rafting into relatively high currents or slightly streptospiral) as it completes the encircling of the possibly by sand delivered through other disturbance pro- proloculus (Figs. 7.1a, b, c). Most specimens are found with cesses. Its peak occurrences in transitional glaciomarine fa- only three chambers (Figs. 7.1, 7.2, 7.3, 7.5, 7.7). If a fourth cies farther south in northernmost Baffin Bay, Lancaster chamber is present, it encircles the test for ½ of a whorl Sound and NE Greenland shelf, but absence from modern in nearly the same plane as the first whorl, but, as for the samples in these areas, suggest that the glaciomarine associ- third chamber, with a tendency to deviate from planispiral ation is pre-eminent, but the specific conditions that produce (Fig. 7.4). Rare specimens with five chambers have been G. oculus peaks are not yet well understood. found (Fig. 7.6). The fifth chamber is added far from the original nearly planispiral plane, making the five- SYSTEMATIC DESCRIPTION chambered specimens look nearly quinqueloculine. A few two-chambered individuals have also been observed. Each Suborder MILIOLINA Delage & Hérouard, 1896 tubular chamber embraces the preceding chambers in an Family FISCHERINIDAE Millett, 1898 evolute chamber arrangement, causing the proloculus to be Subfamily GLOMULININAE Saidova, 1981 visible always, although more of the proloculus is hidden Genus Glomulina Rhumbler, 1936 with the addition of more chambers. In specimens with three chambers, a larger section of the proloculus is visible (Fig. Glomulina oculus n. sp. 7.1) than in specimens with four (Fig. 7.4) and five chambers Figs. 7–11 (Fig. 7.6). The cross-sectional shape of the tubular chambers The subfamily Glomulininae is characterized by a test is partly compressed with the cross section of the tube tend- with proloculus followed by a streptospirally enrolled tubu- ing towards semicircular. Sutures are relatively distinct and lar chamber; later chambers are one-half coil in length. The accentuated by the fact that often the diameter of the tubular genus Glomulina has a globular test and a proloculus fol- chamber decreases lengthwise, but increases distinctly from lowed by a streptospirally enrolled, undivided tubular sec- one chamber to the next, making the tube diameter different ond chamber. Later chambers are one-half coil in length. at each side of the suture. Aperture rounded, at the end of the last chamber. The aperture at the open end of the last tube is rounded, Deriviato Nominis/Deriviation of name. From oculus semicircular in shape and fully to near-fully open (Figs. (Latin) = eye. Refers to the round or eye-like design of spec- 7, 8, 9, 10). Scanning electron microscope images reveal a imens when seen in side view. low thickened bulge or ‘bump’ in the middle of the aper- Diagnosis. Small oblong, globular to discoidal miliolid, tural base, stretching a little from the aperture opening in- normally with a large proloculus followed by 2–4 tubular wards into the last chamber, making the bulge longer than chambers, most commonly 2–3, each of about ¼ to ¾ of wide and it gradually becomes lower inwards into the test, a whorl. The aperture at the open end of the last tubular thus forming a more steep outer end and a less steep inner chamber is open and semicircular. end of the bulge (Fig. 10). This slight bump at the base of Holotype. Three-chambered specimen from OD1507- the aperture is also sometimes visible in light microscopy 23MC-5, 0-2 cm. U.S. National Museum Paleobiology cat- (Figs. 7.2b, 7.5). alog number USNM PAL 726877 is housed at the Smithso- Wall structure is calcareous, imperforate, porcelaneous, nian Institution, Washington, D.C., U.S.A. (Figs. 7.2a, b). but commonly semi-translucent. In scanning electron micro- Paratypes. Thirteen specimens housed at the Smithso- scope (SEM) analyses, a clearly porcellaneous wall structure nian Institution, Washington, D.C., U.S.A. Paratypes 1, 2 (3 (Fig. 11) may be observed with the outer layer (the extra- chambers; USNM PAL 726878 and 726879) and Paratype dos, see Parker, 2017) of aligned calcareous crystal needles 3 (4 chambers; USNM PAL 726880) from OD1507-23MC, primarily ordered lengthwise in direction of the chambers 0–2 cm. Paratypes 4–7 (3 chambers; USNM PAL 726881, (Figs. 11.1, 11.2). Below the extrados, the main body of the 726882, 726883, and 726884) and Paratype 8 (4 cham- wall (porcelain; see Parker, 2017) consists of randomly ori- bers; USNM PAL 726885) from OD1507-45MC-5, 0–1 cm. entated crystal rods, shorter and thicker than those of the Paratypes 9–11 (3 chambers; USNM PAL 726886, 726887, extrados and intrados (Figs. 11.3a, b). Also noteworthy is

Figure 7. Examples of Glomulina oculus from OD1507 multicore (MC) surface sediments (1-6) and a fossil specimen (7) as light microscope photographs. 1a, b side views, 1c apertural view of three-chambered specimen from 47MC showing nearly planispiral chamber arrangement with chambers numbered from proloculus (P), i.e., ﬁrst chamber, and onwards. 2a Side view and 2b apertural view of Holotype (USNM PAL 726877) from 23MC. 3a, b, c Side views of stained three-chambered specimen (live) showing pronounced streptospiral third chamber that obscures the proloculus on one side. 4a, b Side views of four-chambered specimen from 47MC; chamber 2 is obscured in 4b. 5 Apertural view of three-chambered specimen from MC47 showing the small ‘bump’ at the base of the aperture. 6 Paratype (USNM PAL 726889), ﬁve-chambered specimen from 34MC; chamber 2 is not visible. 7 Paratype USNM PAL 726886), fossil specimen of typical three-chambered form from 2008029-59PC, 549-551 cm, All specimens at the same scale. Scale bars = 20 μm.

that in a single specimen, we were able to observe the crystal intrados. The ridges have a swirling pattern, reflecting the structure of the inner layer (intrados; see Parker, 2017) round shape of the proloculus (Figs. 11.6a, b). within the proloculus (Figs. 11.6a, b, c). Here the crystal nee- Despite the fact that the specimens, which were scanned, dles were also randomly oriented, but the crystals were cov- were collected from the upper 2 cm sediment, they already ered by a fine surface layer with fine wavy ridges. This very showed some dissolution of the shell, and in some speci- thin layer inside the proloculus is presumably part of a thin mens the early chambers also showed re-crystallization of organic layer covering the calcareous crystal needles of the the carbonate forming rhomboid crystals at the surface of

Figure 8. Scanning electron microscope photographs (SEM) of Glomolina oculus, all from 0–2 cm depth in cores OD1507-23MC and OD1507- 47MC. 1 Specimen SEM16. 2. Specimen SEM2. 3 Specimen SEM3. 4 Specimens SEM1. 5 Specimen SEM17. 6 Specimen SEM5. 7 Specimen SEM15. 8 Specimen SEM6. 9 Specimen SEM10. 10 Specimen SEM11. 11a, Marginal view of specimen SEM14, note the very large proluculus that is clearly visible on both sides on the 11b view. 12 Specimen SEM12 with proloculus broken on both sides. 13 Specimen SEM4 with a clear view of the inner wall of the proloculus. 13 Specimen SEM13 with proloculus broken on upper side. 14 Specimen SEM8. Scale bar = 50 µm.

Figure 9. Scanning electron microscope (SEM) photographs of the aperture of Glomolina oculus, all from 0-2 cm depth in cores OD1507-23MC and OD1507-47MC. 1a Overview of aperture of specimen SEM8. 1b Detailed view of aperture of Specimen SEM8. 2a, b Specimen SEM2. 3 Specimen SEM3. 4 Specimen SEM1. 5 Specimen SEM5. 6 Specimen SEM16. Scale bar = 10–30 µm.

Figure 10. Scanning electron microscope (SEM) photographs of details of the aperture of Glomolina oculus, all from 0-2 cm depth in cores OD1507-23MC and OD1507-47MC. 1a, b Overview and detailed view, respectively, of aperture of Specimen SEM14. Note the clear bulge at the base of the aperture. 2 Aperture of specimen SEM10, with aperture ﬁlled with sediment and carbonate crystals. 3a, b Apertural view of specimen SEM11. 3c Detailed view of the bump at the base of the aperture of specimen SEM11. Note the wall structure of randomly oriented calcite needles. 4a, b Overview and detail of aperture, respectively, of specimen SEM7. The last part of the last chamber is broken oﬀ providing a side view of the bump at the base of the apertural opening (marked with arrow); the abrupt outer side and tapering internal side of the bump are clearly seen. Scale bar = 5–50 µm.

Figure 11. Scanning electron microscope (SEM) photographs of details of test wall structure Glomolina oculus, all from 0–2 cm depth in cores OD1507-23MC and OD1507-47MC. 1 Broken end of the last chamber of specimen SEM6 showing the thin layer of oriented crystals of the extrados, and at the end the randomly oriented needles the middle section of the test wall (porcelain). 2 The crystals are lengthwise orientated near the aperture. 3a, b Broken test wall showing the randomly oriented crystals of the porcelain (central) part of the wall. 4a, b, 5 Larger calcite crystals seen at the base of the aperture, presumably indicating post-mortem re-crystallization of this older part of the test in specimens SEM12 and SEM17. Note that the upper end of the second chamber of specimen SEM 12 is broken providing a view inside the second chamber. The large crystal seen in the apertural opening of SEM17 is considered a post-mortem feature not relevant for the species description. 6a, b, c Detailed view of the inner wall of the proloculus of specimen SEM 15. 6a, b shows the randomly orientated crystals of the middle (porcelain) section of the wall covered by a thin inner (intrados) wall section with ridges in an irregular, swirly pattern. Scale bar = 5–40 µm.

Figure 12. Scanning electron microscope (SEM) photographs of Cornuspira distincta and a juvenile specimen of Miliolinella subrotunda, all three specimens from 440–441 cm core depth (Marine Isotope Stage 3 in age) in gravity core DA12-11/2-GC03 (60°78238’N, 9°79373’W; 742.4 m water depth) south of the Faeroe Islands. 1a Detail of aperture of C. distincta specimen SEM2-2. 1b C. distincta specimen SEM2-2. 2 C. distincta specimen SEM2-1. 3 M. subrotunda specimen SEM2-3.

the test (Figs. 11.4, 11.5). None of the specimens analyzed in Glomulina oculus has a superficial resemblance to Cornus- the SEM were stained, indicating that they were all dead at pira distincta (Cole & Scott) (originally described as Cyclo- the time of collection. gyra distincta in Scott et al., 2008) that has a similar shape Dimensions. Average test dimensions with details listed in with a large round proloculus followed by a long planispi- Table 1: 161 by 131 µm for three-chambered tests, 196 by 168 ral to slightly streptospiral tube (Fig. 12.1-2). However, G. µm for four-chambered tests, and 242 by 154 µmforfive- oculus is clearly distinguished from C. distincta by having at chambered tests (Table 1). least three chambers, while C. distincta has only two cham- Variation. There is relatively little variation within this bers, and the specimens of C. distincta are generally much species, but the number of chambers varies from two to five, more well-rounded in side view than G. oculus (Fig. 12.3). In including the proloculus (Fig. 7). Three-chambered speci- addition, the aperture of C. distincta is rather triangular with mens are the most common. In specimens with four cham- a basal plate in contrast to the semicircular aperture with- bers, less of the proloculus is exposed on one side, while out any distinct basal plate in G. oculus. The apertural shape in five-chambered specimens, the proloculus is mostly ob- and basal plate of C. distincta is clearly seen in the figures scured. Smaller specimens tend to have a semicircular shape of the original description of C. distincta, and we were able in side view, while the largest specimens are more elliptical to confirm this when investigating specimens under light mi- in shape (side view). croscope and in SEM microphotographs (Fig. 12.1). Among Affinities. Glomulina oculus is distinguished from Glo- the specimens that we studied are those originally described mulina fistulescens Rhumbler (1936), type species of Glo- as Cyclogyrinae by Bergsten (1994) from surface sediment mulina from the Kiel Bight, Germany, by its generally sig- samples from the Yermak Plateau (i.e., the same material nificantly fewer chambers (none of the specimens depicted that contained the specimens assigned as lectotypes for C. inthetypedescriptionofGlomulina fistulescens have less distincta by Scott et al., 2008). We have not been able to test than five chambers, while 5-chambered G. oculus are rare), the specimens of Schröder-Adams et al. (1990) from east- resulting in a flatter shape of the G. oculus test. Glomulina ern Canadian Arctic shelf sediments, but the photographed oculus differs from Glomulina rotiensis Loeblich & Tappan specimens clearly look like C. distincta and were also as- (1994) from the western Timor Sea by its rounder aperture signed as lectotype for this species by Scott et al. (2008). It and its more distinct suture between the proloculus and the is noteworthy that the genus Cornuspira does not have an 2–3 surrounding chambers, giving G. oculus a more uneven, apertural plate, as defined by Schultze (1854), and thus C. flattened quatrefoil shape in peripheral view (Figs. 7.1c, distincta should potentially be referred to a different genus. 8.10, 8.11b), than the more smoothly rounded Glomulina Glomulina oculus may also express some similarity to ju- rotiensis. venile specimens of Miliolinella subrotunda (Montagu), as

Table 1. Measurements of the dimensions of specimens of three, four, and ﬁve-chambered Glomulina oculus including the range and averages (in brackets) of length and width of the tests and the range of the diameters of the proloculus and the aperture.

Specimen type orfeature No. of measurements Length (average)(µm) Width (average)(µm) Diameter(µm)

three-chambered 15 134–213 (161) 103–192 (131) - four-chambered 6 169–213 (196) 159–175 (168) - ﬁve-chambered 2 239–244 (242) 150–159 (154) - Proloculus 22 - - 53–111 Aperture 7 - - 24–37

some specimens of M. subrotunda do have a large prolocu- generally low preservation potential of G. oculus might ex- lus. However, M. subrotunda may be distinguished from G. plain why these thin-walled species have not been observed oculus by its quinqueloculine chamber arrangement and a previously, a fact, which is surprising considering its rela- large apertural basal plate, although the latter character can tively high frequencies in some of our samples. However, it be difficult to identify in juvenile specimens (see Fig. 12.3). is more likely that it has hitherto been included in counts of Remarks. Only specimens with a large proloculus were other species such as Cornuspira distincta and juvenile spec- found. Whether this means that only megalospheric (hap- imens of Miliolinella subrotunda. loid) individuals of Glomulina oculus were found, as recog- nized for other species with numerous successive asexual cy- CONCLUSIONS cles (cf., sen Gupta, 1999), or whether there is no differ- ence between haploid and diploid generations is yet uncer- Glomulina oculus n. sp. lives today in the northern Nares tain. The fact that some specimens have been found with five Strait and the Petermann Fjord (NW Greenland), as well as chambers, while the majority of specimens have three cham- off Zachariae Isbrae (NE Greenland) in relation to a strat- bers and the second-most common form has four chambers ified water column with chilled and modified Atlantic Wa- is somewhat puzzling. One reason may be that the three- ter underlying Polar Water. It typically occurs in an environ- and four-chambered individuals are juveniles that because ment with marine terminating glaciers and seasonally mo- of unfavorable environmental conditions rarely reach matu- bile sea-ice cover that allows episodic marine productivity, rity (five chambers), but it is also possible that the species and it has an affinity for increased sand content, although is developing towards one that will reproduce in a more ju- further studies are needed to establish its full range of habi- venile form (netenious species). Retardation of growth may tats. In sediment cores, Glomulina oculus has been found ex- be related to lack of food (cf., Boltovskoy et al. 1991). It clusively in early deglacial glaciomarine sediments of north- is also possible, however, that the normal ‘adult’ form of ern Baffin Bay and in similar settings on the NE Greenland the species is in fact three (or four) chambers and that the shelf. The species has been shown to be reasonably well pre- five-chambered individuals are the ones that, due to harsh served in both freeze-dried samples and in samples prepared living conditions, are not able to reproduce at its regular and counted wet, suggesting that the lack of previous reg- stage and keep growing for longer time (cf., Boltovskoy et al. istration is presumably not related to a poor preservation 1991). The question on a possible relation between repro- state, but rather to its specific environmental requirements duction cycle, ontogeny and ecology was also pointed out and likely also to misidentification in previous studies due by Boltovskoy and Wright (1976) but listed as one of the ‘un- to a resemblance to Cornuspira distincta and juvenile speci- resolved problems’. Further data on the distribution of the mens of Miliolinella subrotunda. different forms of Glomulina oculus in relation to environmental conditions is needed to help in resolving this issue. ACKNOWLEDGMENTS

PRESERVATION POTENTIAL AND LIKELIHOOD We are grateful to the captain and crew of Oden and the OF FINDING GLOMULINA OCULUS IN MODERN Oregon State University Coring team, for recovering the AND FOSSIL SAMPLES multicores and to Maziet Cheseby and Laurence Dyke for sampling them on deck. We thank Alan Mix and the scien- The preservation potential for Miliolina is generally low tific party of the 2015 Petermann Glacier Expedition for pro- because of the high content of Mg in the shells (cf., Brown viding the samples. We thank the Atlantic Geoscience Cen- & Elderfield, 1996; sen Gupta, 1999; Petró et al., 2018). The ter, Dartmouth, Canada for access to 2001LSSL-14PC and specimens of Glomulina oculus that we have encountered so 2008029-49 and -59P. We are also grateful to the captain, far are all very small and thin-shelled. They tend to break crew, and scientific party of the NorthGreen2017 expedition very easily, when handled; in particular, the proloculus was that helped retrieve cores DA17-NG-ST07-73G and DA17- broken in many of the specimens (Fig. 8). Even specimens NG-ST08-90R. that in light microscope seemed very well-preserved proved, We are furthermore grateful to Electron Microscope when analyzed in scanning electron microscope, to have been (EM) Engineer Ramon Liebrechts from the Interdisci- subject to early stages of dissolution and diagenesis, most plinary Nanoscience Center (iNANO) at Aarhus Univer- commonly decomposing the outermost thin organic layer sity, who provided invaluable help during scanning elec- of the test material to reveal the extrados layer of calcare- tron microscope (SEM) photomicrographing. We would ous needle-shaped crystals. Some specimens also showed re- also like to thank Professor Kjell Nordberg, University of crystallization and, likely post-mortem, growth of rhombic Gothenburg, Sweden, and Dr. Helena Bergsten for placing calcite crystals in the older parts of the test as well as within their specimens of “Cyclogyrinae” from Bergsten (1994) at the aperture. Also, the thin wall of the proloculus as of- our disposal for comparison and to Dr. Teodóra Pados- ten broken on one or both sides. However, it is notewor- Dibattista and Joanna M. Davies Aarhus University, for the thy that we did indeed find numerous specimens of G. ocu- foraminiferal analyses of the NE Greenland cores. Many lus in early Holocene sediments at several meters depth in thanks to Dr. Kelly Hogan, British Antarctic Survey, for the cores. Also, the specimens have been subjected to both making the map in Figure 1B. We thank Mark Florence freeze-drying and oven-drying at low temperatures (25°C) of the Smithsonian Natural History Museum for providing during the laboratory process. This indicates that the preser- the museum numbers for the Holotype and paratypes. We vation potential is not as low as could be assumed from their also thank the journal editors and two anonymous review- very thin-walled, fragile porcelaneous tests. Nevertheless, a ers for their comments. Funding for Anne Jennings’ research

was provided by National Science Foundation PLR-ANS Jennings, A. E., Andrews, J. T., Oliver, B., Walczak, M., and Mix, 1417784, “Collaborative Research: Petermann Gletscher, A., 2019, Retreat of the Smith Sound Ice Stream in the Early Holocene: Boreas, v. 48, p. 825-840. Greenland - Paleoceanography and Paleoclimatology” and Jennings, A. E., and Helgadóttir, G., 1994, Foraminiferal assem- OPP 1804504 “Timing and Paleoceanographic Impacts of blages from the fjords and shelf of eastern Greenland: Journal of the Onset of Arctic-Baffin Bay Throughflow”. Marit-Solveig Foraminiferal Research, v. 24, p. 123–144. Seidenkrantz was funded by the Independent Research Kozo, T. L., 1991, The hybrid polynya at the northern end of Nares Fund Denmark / Natural Science (G-Ice project grant no Strait: Geophysical Research Letters, v. 18, p. 2059–2062. Kwok, R., 2005, Variability of Nares Strait ice flux: Geophysical Re- 7014-00113B/FNU), and through the NorthGreen2017 ex- search Letters, v. 32, doi: 10.1029/2005GL024768. pedition grant from the Danish Centre for Marine Research. Kwok, R., Toudal Pederson, L., Gudmandsen, P., and Pang, S.S., 2010, Large sea ice outflow into the Nares Strait in 2007: Geophysical Research Letters, v. 37, doi: 10.1029/2009GL041872. REFERENCES Larsen, N. K., Levy, L. B., Carlson, A. E., Buizert, C., Olsen, J., Strunk, Aagaard, K., and Coachman, L. K., 1968a, The East Greenland Cur- A., Bjørk, A. A., and Skov, D. S., 2018, Instability of the Northeast rent north of Denmark Strait, Part I: Arctic, v. 21, p. 181–200. Greenland Ice Stream over the last 45,000 years. Nature Commu- Aagaard, K., and Coachman, L. K., 1968b, The East Greenland Cur- nications 9, 1872. Doi:10.1038/s41467-018-04312-7. rent north of Denmark Strait, Part II: Arctic, v. 21, p. 267–290. Lloyd, J. M., 2006, Modern distribution of benthic foraminifera from Bergsten, H., 1994, Recent benthic foraminifera of a transect from the Disko Bugt, West Greenland: Journal of Foraminiferal Research, North Pole to the Yermak Plateau, eastern central Arctic Ocean: v. 36, p. 315–331. Marine Geology, v. 119, p. 251–267. Loeblich, A. R., Jr., and Tappan, H., 1994, Foraminifera of the Sahul Boltovskoy, E., and Wright, R., 1976, Recent Foraminifera: Dr. W. Shelf and Timor Sea: Cushman Foundation for Foraminiferal Re- Junk, The Hague, 515 p. search, Special Publication, No. 31, 661 p. Boltovskoy, E., Scott, D. B., and Medioli, F. S., 1991, Morphological Makinson, K., and Anker, P. G., 2014, The BAS ice-shelf hot-water variations of benthic foraminiferal tests in response to changes in drill: Design, methods and tools: Annals of Glaciology, v. 55, ecological parameters: A review. Journal of Paleontology, v. 65, p. 44–52. p. 175-185. Melling, H., Gratton, Y., and Ingram, G., 2001, Ocean circulation Brown, S., and Elderfield, H., 1996, Variations in Mg/Ca and Sr/Ca ra- within the North Water polynya of Baffin Bay: Atmosphere-Ocean, tios of planktonic foraminifera caused by postdepositional dissolu- v. 39, p. 301–325. tion: Evidence of shallow Mg-dependent dissolution: Paleoceanog- Millet, F. W., 1898, Report on the Recent foraminifera of the Malay raphy, v. 11, p. 543–551. Archipelago collected by Mr. A. Durrand, F. R. M. S., Part III: Darling, K. F., Schweizer, M., Knudsen, K. L., Evans, K. M., Bird, Journal of the Royal Microscopical Society, p. 607-614. C., Roberts, A., Filipsson, H., Kim, H.-J., Gudmundsson, G., Muenchow, A., 2019, Physical ocean property profile from Petermann Wade, C. M., Sayer, M. D. J., and Austin, W. E. N., 2016: The Fjord, Greenland, surveyed in 2015 by IB/Oden. Arctic Data Cen- genetic diversity, phylogeography and morphology of Elphidiidae ter. doi:10.18739/A2XS5JH16. (Foraminifera) in the Northeast Atlantic: Marine Micropalaeon- Münchow, A., Falkner, K. K., Melling, H., Rabe, B., and John- tology, v. 129, p. 1–23. son, H. L., 2011, Ocean warming of Nares Strait bottom wa- Delage, Y., and Herouard, E., 1896, Traité de Zoologie Concrète, Vol ters off Northwest Greenland, 2003–2009: Oceanography, v. 24, 1, La Cellule et les Protozoaires: Schleicher Frères, Paris, 618 p. p. 114–123. Dunbar, M., 1969, The geophysical position of the North Water: Arctic, Münchow, A., Padman, L., Washam, P., and Nicholls, K. W., 2016, v. 22, p. 438–441. The ice shelf of Petermann Gletscher, North Greenland, and its Ellis, B. F., and Messina, A., 1949 (and update supplements), Cata- connection to the Arctic and Atlantic Oceans: Oceanography, logue of Foraminifera: American Museum of Natural History and v. 29, p. 84–95. Micropalaeontology Press, New York (ISBN 978-0-913424-34-6), Osterman, L. E., Poore, R. Z., and Foley, K. M., 1999, Distribu- http://wwwmicropress.org/em/about.php (Accessed 22-06-2019). tion of benthic foraminifers (>125 μm) in the surface sediments Gooday, A. J., and Hughes, J. A., 2002, Foraminifera associated with of the Arctic Ocean: US Geological Survey Bulletin no. 2164, phytodetritus deposits at a bathyal site in the northern Rockall p. 1–28. Trough (NE Atlantic): seasonal contrasts and a comparison of Parker, J. H., 2017, Ultrastructure of the test wall in mod- stained and dead assemblages: Marine Micropaleontology, v. 46, ern porcelaneous foraminifera: implication for the classifica- p. 83–110. tion of the Miliolida: Journal of Foraminiferal Research, v. 47, Hald, M., and Korsun, S., 1997, Distribution of modern benthic p. 136-174. foraminifera from fjords of Svalbard, European Arctic: Journal of Petró, S. M., Pivel, M. A. G., and Coimbra, J. C., 2018, Foraminiferal Foraminiferal Research, v. 27, p. 101-122. solubility rankings: A contribution to the search for consensus: Heuzé, C., Wåhlin, A., Johnson, H. L., and Münchow, A., 2017, Path- Journal of Foraminiferal Research, v. 48, p. 301–313. ways of meltwater export from Petermann Glacier, Greenland: Reilly, B. T., Stoner, J. S., Mix, A. C., Walczak, M. H., Jennings, A. E., Journal of Physical Oceanography, v. 47, p. 405–418. Jakobsson, M., Dyke, L., Glueder, A., Nicholls, K., Hogan, K. A., Jakobsson, M., Hogan, K. A., Mayer, L. A., Mix, A., Jennings, A., Mayer, L. A., Hatfield, R. G., Albert, S, Marcott, S., Fallon, S., Stoner, J., Eriksson, B., Jerram, K., Mohammad, R., Pearce, C., and Cheseby, M., 2019, Holocene break-up and reestablishment Reilly, B., and Stranne, C., 2018, The Holocene retreat dynamics of the Petermann Ice Tongue, northwest Greenland: Quaternary and stability of Petermann Glacier in northwest Greenland: Na- Science Reviews, v. 218, p. 322-342. ture Communications, v. 9, https://doi.org/10.1038/s41467-018- Rhumbler, L., 1936, Foraminiferen der Kieler Bucht, gesammelt durch 04573-2. A. Remane, Teil II – Ammodisculinidae bis einschluss Textular- Jennings, A., and Andrews, J., 2019, Petermann Expedition, Oden inidae: Kieler Meeresforschungen, v.1, p. 179–242. July 2015, marine core data for mineralogy, foraminifera Ryan, P. A., and Münchow, A., 2017, Sea ice draft observations in and particle size, Greenland, 2019: Arctic Data Center, doi: Nares Strait from 2003 to 2012: Journal of Geophysical Research, 10.18739/A2MP4VN1H. Oceans, v. 122, p. 3057–3080. Jennings, A. E., Weiner, N. J., Helgadottir, G., and Andrews, J. T., 2004, Saidova, K. M., 1981, On an up-to-date system of supraspecific Modern foraminiferal faunas of the southwestern to northern Ice- taxonomy of Cenozoic benthonic foraminifera. Moscov: Insti- land shelf: Oceanographic and environmental controls: Journal of tut Okeanologi P. P. Shirshova, Akadamia Nauk, SSSR. 73 p Foraminiferal Research, v. 34, p. 180–207. (in Russian). Jennings, A. E., Sheldon, C., Cronin, T. M., Francus, P., Stoner, J., and Samelson, R. M., Agnew, T., Melling, H., and Münchow, A., 2006, Andrews. J., 2011, The Holocene history of Nares Strait: Transi- Evidence for atmospheric control of sea-ice motion through tion from glacial bay to Arctic-Atlantic throughflow: Oceanogra- Nares Strait: Geophysical Research Letters, v. 33, L02506, phy, v. 24, p. 26–41. doi: 10.1029/2005GL025016.

Schröder-Adams, C. J., Cole, F. E., Medioli, F. S., Mudie, P. J., Scott, Tang, C., Ross, C., Yao, T., Petrie, B., Detracey, B., and Dunlap, E., D. B., and Dobbin, L., 1990, Recent Arctic shelf foraminifera: 2004, The circulation, water masses and sea-ice of Baﬃn Bay: Seasonally ice covered vs. perennial ice covered areas: Journal of Progress in Oceanography, v. 63, p. 183–228. Foraminiferal Research, v. 20, p. 8–36. Washam, P., Münchow, A., and Nicholls, K. W., 2018, A decade of Schultze, M. S., 1854, Über den Organismus der Polythalamien ocean changes impacting the ice shelf of Petermann Gletscher, (Foraminiferen), nebst Bemerkungen über die Rhizopoden im Al- Greenland: Journal of Physical Oceanography, v. 48, p. 2477–2493. gemeinen: Wilhelm Engelmann, Leipzig, 68 p. Washam, P., Nicholls, P. W., Münchow, A., and Padman, L., Scott, D. B., Schell, T., Rochon, A., and Blasco, S., 2008, Modern ben- 2019, Summer surface melt thins Petermann Gletscher Ice Shelf thic foraminifera in the surface sediments of the Beaufort shelf, by enhancing channelized basal melt: Journal of Glaciology, slope and Mackenzie Trough, Beaufort Sea, Canada: Taxonomy doi: 10.1017/jog.2019.43. and summary of surﬁcial distribution: Journal of Foraminiferal Wollenburg, J. E., and Mackensen, A., 1998, Living benthic Research, v. 38, p. 228–250. foraminifera from the central Arctic Ocean: Faunal composition, Seidenkrantz, M.-S., 1995, Cassidulina teretis Tappan and Cassidulina standing stock and diversity: Marine Micropaleontology, v. 34, neoteretis new species (Foraminifera): Stratigraphic markers for p. 153–185. deep sea and outer shelf areas: Journal of Micropalaeontology, v. 14, p. 145–157. sen Gupta, B. K., 1999, Modern Foraminifera: Kluwer, Dordrecht, The Received 9 August 2019 Netherlands, 371 p. Accepted 15 February 2020

Downloaded from https://pubs.geoscienceworld.org/cushmanfoundation/jfr/article-pdf/50/2/219/5092044/i0096-1191-50-2-219.pdf by University of Colorado Boulder user on 21 July 2020